Patents
WO 2026/068427 A1: Method for achieving relative encoding of a signal
Guillaume Larue, Louis-Adrien Dufrène, Quentin Lampin
April 02, 2026 — International Patent Application
Disclosed is a method for encoding a source message via a symbol frame, the source message containing a first number (l) of initial variables $(x_0, x_1, ..., x_{l-1})$, each initial variable being discrete and taking values among a finite set (F), the method comprising the steps of: - determining a second number (k) of relative variables $(y_0, y_1, ..., y_{k-1})$ expressed by a change of variable using the initial variables $(x_0, x_1, ..., x_{l-1})$, - encoding the relative variables $(y_0, y_1, ..., y_{k-1})$ using a predefined coding matrix (G), said encoding resulting in a third number (n) of coded relative variables $(z_0, z_1, ..., z_{n-1})$, the third number (n) being greater than the second number (k), - determining a fourth number (m) of coded absolute variables $(c_0, c_1, ..., c_{m-1})$ expressed by the change of variable using the coded relative variables $(z_0, z_1, ..., z_{n-1})$, and in which the symbol frame contains said coded absolute variables $(c_0, c_1, ..., c_{m-1})$.
WO 2026/068108 A1: Method for achieving relative decoding of an ofdm frame
Guillaume Larue, Louis-Adrien Dufrène, Quentin Lampin
April 02, 2026 — International Patent Application
The invention relates to a method for decoding a coded frame at the output of a transmission channel, said coded frame comprising coded symbols including encoding on data relating to relative offsets between initial variables, the method comprising the following steps: - obtaining, from the coded frame, a tensor of relative distributions (P) comprising a plurality of relative distributions (Pi/i') representing probabilities of relative deviations between two given coded symbols (i, i'), - obtaining coding relationships (S) linking the coded symbols and translating relative relationships between the coded symbols, and supplementing them based on at least one symmetry and/or transitivity relationship between the relative distributions, - based on the coding relationships thus supplemented, expressing relative relationships between the relative distributions, - determining an estimate of decoded symbols via a probabilistic decoding processing operation using the coding relationships (S).
WO 2026/017825 A1: Management of the transmission and reception of a compressed data packet
Quentin Lampin, Guillaume Larue, Louis-Adrien Dufrène
January 22, 2026 — International Patent Application
The invention relates to a method for managing the transmission of a data packet compressed by applying a compression rule belonging to a set of compression rules, referred to as the compressed packet (P), comprising the insertion into the compressed packet (P) of an identifier of the set of compression rules, referred to as the unique context identifier (ICU), followed by the transmission of the thus modified compressed packet (P').
WO 2026/017824 A1: Rule-based compression method and corresponding management entity
Quentin Lampin, Louis-Adrien Dufrène, Guillaume Larue
January 22, 2026 — International Patent Application
The invention relates to a method for determining an identifier of a set of data compression rules, a rule comprising at least one target value of at least one data field, said method being implemented by a data processing unit and comprising: applying a transformation to at least one target value selected from the target values included in the rules of the set of compression rules, the determined identifier resulting from said application.
WO 2026/002906 A1: Method for optimizing a linear code
Louis-Adrien Dufrène, Guillaume Larue, Quentin Lampin
January 02, 2026 — International Patent Application
The invention relates to a method for optimizing a function (F), the function (F) being a linear code with which a control matrix is associated, the function (F) comprising parameters that are the elements of the control matrix, the parameters taking their values from a set (E) of discrete values, the method using another function (L), referred to as a loss function, the method, which is implemented by a data processing device (100), comprising: executing a gradient descent step on the loss function (L) in order to obtain a parameter value (V1, V2, V3), the obtained values optimizing the function (F).
WO 2025/219411 A1: Method for managing the communication of an item of vector data
Quentin Lampin, Guillaume Larue, Louis-Adrien Dufrène
October 23, 2025 — International Patent Application
The invention relates to a method for managing the communication of an item of vector data (SD), referred to as the item of data to be communicated, from a first device (EQ1) to a second device (EQ2), the method comprising: obtaining an index (I') by applying an indexing function (F) to the item of data to be communicated (SD), wherein the indexing function (F) is such that the closer two items of vector data are, as measured by a distance (D) between them, the closer the indices calculated for these two items of vector data by the function (F) are to one another, wherein the index (I') makes it possible to retrieve, in a data structure (DS), another item of vector data (SD'), referred to as the item of communicated data, wherein the data structure (DS) associates indices with items of vector data, and wherein the index (I') is capable of being transmitted from the first device (EQ1) to the second device (EQ2).
WO 2025/132969 A1: Methods, Devices and Programs for Demodulating a Data Frame
Guillaume Larue, Louis-Adrien Dufrène, Quentin Lampin
June 26, 2025 — International Patent Application
The invention relates to a method for demodulating a data frame comprising N information words $[P_0, ..., P_{N-1}]$, in which a word $p_i$ occupying a position $i \in [0, N-1]$ in the frame is modulated by a symbol $C_{p_i}$ from a dictionary of $K$ symbols, wherein the method comprises the steps of determining (301), for a plurality of symbols $C_{p_i}$ and $C_{p_j}$ of the frame, where $i,j \in [0, N-1]$, a vector of $K$ probabilities $P_{i/j}$, referred to as the relative distribution, comprising probabilities associated with $K$ possible values of an estimated difference between the symbols $C_{p_i}$ and $C_{p_j}$; combining (303) a relative distribution $P_{i/j}$ with a particular probability distribution mask $i/j$ determined (302) according to a characteristic of the channel; and demodulating (304) the frame on the basis of at least one portion of the combined distributions.
WO 2025/132968 A1: Methods, Devices and Programs for Adaptive Modulation and Demodulation of a Data Frame
Guillaume Larue, Louis-Adrien Dufrène, Quentin Lampin
June 26, 2025 — International Patent Application
The invention relates to a method for demodulating a frame comprising N information words $[P_0, ..., P_{N-1}]$, in which a word $P_i$ at a position $i \in [0, N-1]$ in the frame is modulated by a symbol $C_{p_i}$ from a dictionary of $K$ symbols. The method is characterised in that it comprises steps of: obtaining (300) a priori knowledge about the distribution of symbols at a plurality of positions in the frame; determining (301) a differential representation of the a priori knowledge; determining (303) a differential representation of the received frame; linking (304) the determined differential representations in order to obtain a consolidated estimate of the differences between symbols in the received frame; and demodulating (305) the information words received from the consolidated estimate obtained. The invention also relates to a corresponding modulation method and to devices and programs for implementing the methods.
WO 2025/132966 A1: Method, device and program for demodulating a data frame
Guillaume Larue, Louis-Adrien Dufrène, Quentin Lampin
June 26, 2025 — International Patent Application
The invention relates to a method for demodulating a signal transmitted over a transmission channel on which a frame of binary words modulated by a particular symbol from a symbol dictionary is received (200), the method comprising steps of determining (201) a set of relative distributions representative of a difference between symbols and the frame; determining (202), for each relative distribution of the set, a partial relative distribution by relating a first relative distribution determined from symbols and with a second relative distribution determined from a symbol and the symbol; consolidating (203) one of the relative distribution by recombination of the determined partial relative distributions; and demodulating (205) the frame from at least a portion of the set of consolidated relative distributions.
US 2025/0199761 A1: Product Reduction Tree for Machine Representation
Guillaume Larue, Louis-Adrien Dufrène, Quentin Lampin
June 19, 2025 — US Patent Application
A computing means which aggregates initial distributions into an aggregated distribution, the initial distributions being associated with vectors of real values representable in a predefined representation format in which the values representable in this format are between a minimum representable value and a maximum representable value, where, for the aggregation of the initial distributions, the computing means is configured to implement, iteratively: a grouping of distributions to form a plurality of distribution subsets, and product operations, term by term, on the distribution subsets, the products resulting in at least one intermediate distribution, followed by a normalization operation on each intermediate distribution, until the aggregated distribution is obtained, where each grouping of distributions is implemented such that the product operations on the distribution subsets lead to resulting values that are strictly greater than the minimum representable value.
US 2024/0064045 A1: Methods and Devices for Freezing or Adapting Parameters of a Neural Network which are used in a Telecommunication Network
Louis-Adrien Dufrène, Quentin Lampin, Guillaume Larue
February 22, 2024 — US Patent Application
In a communication network, an item of equipment uses a first neural network to implement a signal processing function in order to process an input signal received over a communication channel to obtain an output signal. A system for adapting the parameters of the first network, after a change in the channel, sends information items about the change for processing by a second neural network trained in association with the first neural network and used to determine parameters of the first network that are to be frozen and parameters of the first network that are to be adapted following a change in the channel. The system also obtains information items supplied by the second neural network which identify the determined parameters of the first network to be frozen and adapted following the detected change, and adapts the identified parameters of the first neural network using the supplied information items.
US 2023/0351156 A1: Method and System for Adapting a Neural Network Used in a Telecommunication Network
Quentin Lampin, Louis-Adrien Dufrène, Guillaume Larue
November 02, 2023 — US Patent Application
In a communication network, an item of equipment uses a first neural network to implement a signal processing function, in order to process an input signal to obtain an output signal. A third neural network configured to determine a transfer function for transferring the parameters of a second neural network to the parameters of the first neural network is trained, the second neural network being less complex than the first network and also being used to implement the processing function, the first and second neural networks having been trained by the same input and output signals. The transfer function allows the parameters of the first network to be deduced from parameters of the second neural network. After detection of a change in the processing function, the parameters of the second network are adapted by means of input signals associated with a training sequence, and the parameters of the first neural network are adapted by using the adapted parameters of the second network and the transfer function.